CN114281833A - Method for rapidly keeping consistency of slave computer and host computer data under master-slave copy mode of OpenGauss database - Google Patents
Method for rapidly keeping consistency of slave computer and host computer data under master-slave copy mode of OpenGauss database Download PDFInfo
- Publication number
- CN114281833A CN114281833A CN202111592861.XA CN202111592861A CN114281833A CN 114281833 A CN114281833 A CN 114281833A CN 202111592861 A CN202111592861 A CN 202111592861A CN 114281833 A CN114281833 A CN 114281833A
- Authority
- CN
- China
- Prior art keywords
- wal
- transaction
- thread
- playback
- logs
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 38
- 230000010076 replication Effects 0.000 claims abstract description 8
- 238000002407 reforming Methods 0.000 claims abstract description 3
- 101100257624 Arabidopsis thaliana SPS4 gene Proteins 0.000 claims description 11
- 238000012545 processing Methods 0.000 claims description 8
- 230000008859 change Effects 0.000 claims description 6
- 238000004806 packaging method and process Methods 0.000 claims description 6
- 238000013507 mapping Methods 0.000 claims description 5
- 238000003672 processing method Methods 0.000 claims description 5
- 230000008569 process Effects 0.000 claims description 4
- 238000009825 accumulation Methods 0.000 abstract description 4
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000007547 defect Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 238000009530 blood pressure measurement Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000003780 insertion Methods 0.000 description 1
- 230000037431 insertion Effects 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
Images
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention belongs to the technical field of relational database management and operating systems, and particularly relates to a method for rapidly keeping the consistency of slave computers and host computer data in a master-slave replication mode of an OpenGauss database and application thereof. The method comprises the steps of reforming the existing WAL playback thread, splitting independent thread playback into multi-thread parallel playback of backup library playback thread work, dividing WAL types into transaction logs and non-transaction logs, adopting independent thread playback aiming at the transaction logs to avoid out-of-order events, and adopting multi-thread parallel playback aiming at the non-transaction logs to accelerate WAL playback speed. By the method, WAL playback of the standby library can be quickly finished, WAL accumulation is reduced, the difference value between the WAL of the main library and the WAL of the standby library is always kept within a set threshold range, the external service capability of the OpenGauss database in a master-slave copy mode is remarkably improved, and technical support is provided for normal performance of the database.
Description
Technical Field
The invention belongs to the technical field of relational database management and operating systems, and particularly relates to a method for rapidly keeping the consistency of slave computers and host computer data in a master-slave replication mode of an OpenGauss database and application thereof.
Background
OpenGauss is an open-source relational database management system, the kernel of which is derived from PostgreSQL, and the system is used as an open-source free database platform and aims to encourage community contribution and cooperation. Currently, OpenGauss databases already support stream-oriented physical replication (stream replication) of a pre-written log (WAL), and database users can receive the WAL from a main library and play back the received WAL during a standby library process through the stream replication, and the standby library continuously receives and plays back the WAL from the main library to complete construction of the standby library and data synchronization. However, as the pressure of the master library increases to generate a large amount of WAL, the WAL accumulation is caused by the slow playback speed of the WAL of the backup library, and the performance of the database is obviously reduced after the difference value between the WAL of the master library and the WAL of the backup library reaches a set threshold value. Therefore, if a technical scheme is provided, the backup library can quickly complete WAL playback, WAL accumulation is reduced, and the difference value between the WAL of the master library and the WAL of the backup library is always kept within a set threshold range, the external service capability of the OpenGauss database in a master-slave copy mode can be obviously improved, and a powerful technical guarantee is provided for normal performance of the database.
Disclosure of Invention
The invention provides a solution for overcoming the defect that the performance of a database is reduced due to inconsistency of slave computer data and host computer data caused by slow playback speed of a standby library WAL in a master-slave copy mode of an OpenGauss database. The invention aims to design a method for rapidly keeping the consistency of slave computers and host computer data in a master-slave copy mode for an OpenGauss database.
Specifically, the invention provides a method for rapidly keeping the consistency of slave computers and host computer data in a master-slave replication mode of an OpenGauss database, which comprises the steps of reforming the existing WAL playback thread, splitting the work of a backup library playback thread into multi-thread parallel playback from single thread playback, and distinguishing WAL types into transaction logs and non-transaction logs, wherein the transaction logs comprise transaction operation related logs, and the non-transaction logs comprise data operation related logs; aiming at the transaction log, the method still adopts independent thread playback to avoid the occurrence of out-of-order events, and aiming at the non-transaction log, the method adopts multi-thread parallel playback to accelerate the WAL playback speed.
Further, the transaction log in the method of the present invention includes a csn (commit SEQUENCE num) log and a transaction commit log; the non-transaction logs include insert operation logs and update operation logs.
Furthermore, in the implementation process of the method of the present invention, when the non-transaction logs are processed by multithread parallel playback, if there is a dependency relationship between the non-transaction logs, the dependency relationship is distinguished as an intra-transaction dependency relationship and an inter-transaction dependency relationship, and for different types of dependency relationships, the following different processing methods are adopted:
(1) aiming at the intra-transaction dependency relationship, when a WAL which operates the same operation object for multiple times exists in one transaction, the number of a working thread for replaying the log is calculated according to a digital label which is obtained from WAL information and used for identifying the physical position of the operation object during parallel replay, and the digital label for identifying the physical position of the operation object has uniqueness, so that all non-transaction logs containing the same operation object are replayed by the same working thread;
(2) and packaging WALs played back in parallel aiming at the inter-transaction dependency relationship, storing related marking information in a packaging structure, setting WAL types and dispatching function mapping relationships in a database kernel, calling the matched dispatching functions aiming at different types of WALs, and processing the inter-transaction dependency relationship by the dispatching functions.
Further, the processing method (2) calls the dispatch function matched with the WAL for different types in the inter-transaction dependency relationship, and the dispatch function processes the inter-transaction dependency relationship, specifically including the following processing modes:
(1) when the type of the WAL is RM _ HEAP _ ID, the corresponding dispatch function is dispatchheapcord, which decides by which REDO _ BATCH thread the WAL is played back according to whether the WAL involves a data block change:
(a) if max _ bloc _ id > is 0, indicating that the data blocks are changed, traversing each data block, selecting grouping according to the physical file OID of the data block, and ensuring that the WALs same as the physical file OID are played back by the same working thread;
(b) if max _ block _ id <0, indicating that no data block change is involved, allocating the WAL to a plurality of working threads for parallel playback;
(2) when the type of the WAL is RM _ STANDBY _ ID, the corresponding dispatch function is DispatchStandbyRecord which dispatches the WAL to a REDO _ TRXN _ MNG thread for playback;
(3) when the type of the WAL is RM _ XACT _ ID, the corresponding dispatch function is DispatcXactRecord, which first determines whether the WAL will delete the file:
(a) if the file is deleted, the WAL is allocated to all REDO _ BATCH threads for playback;
(b) if the file is not deleted, the WAL is assigned to the REDO _ TRXN _ MNG thread for playback.
Preferably, the method for rapidly keeping the slave computer consistent with the host computer data in the master-slave copy mode of the OpenGauss database in the present invention includes the following steps:
(1) the log receiving thread writes the WAL sent by the master library into a disk;
(2) reading the WAL from the disk by the Start thread, and distinguishing the WAL into a transaction log and a non-transaction log, wherein:
(a) the transaction log is played back by the REDO TRXN work thread alone,
(b) non-transactional logs are replayed in parallel by multiple threads;
(3) for the non-transaction logs, different dispatching functions are called according to the types of the non-transaction logs, the dispatching functions dispatch the non-transaction logs and place the non-transaction logs into SPSC queues of corresponding REDO _ PAGE _ WORKER threads, and if the queues are full, the start threads wait circularly;
(4) the REDO _ PAGE _ WORKER thread reads the non-transaction log from the SPSC queue and plays back the non-transaction log, and updates the position of the non-transaction log played back by the thread;
(5) and acquiring the minimum value of the non-transaction logs played back by the REDO _ PAGE _ WORKER thread by the Start thread, judging whether the transaction logs can be played back or not, and calling a corresponding playback function if the transaction logs can be played back.
In addition, the invention also relates to the application of the method for rapidly keeping the consistency of the slave computer and the host computer data in the master-slave copy mode of the OpenGauss database in a relational database management or operating system.
In summary, the invention provides a method for rapidly keeping the consistency of slave computer data and host computer data in the master-slave copy mode of the OpenGauss database for the first time, and the method well overcomes the defect that the slave computer data and the host computer data are inconsistent and the performance of the database is reduced due to the slow playback speed of the slave computer WAL in the master-slave copy mode of the OpenGauss database by designing a set of WAL parallel playback scheme. The WAL playback of the standby library can be rapidly finished by using the method, the WAL accumulation is reduced, the difference value between the WAL of the main library and the WAL of the standby library is always kept within a set threshold range, and the result shows that the performance of the database after the WAL parallel playback is started is obviously superior to that in an unopened state through pressure measurement in a master-slave mode.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the embodiments of the present invention are briefly described below, it is obvious that the following drawings are only some embodiments described in the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a schematic flow chart of the method of the present invention.
Drawing notes: start thread: WAL processing thread;
SPSC: for saving the WAL queue;
REDO _ BATCH thread: reading and analyzing WAL from the SPSC queue of the start thread, and then putting the WAL into the SPSC queue of the REDO _ PAGE _ MNG thread;
REDO _ PAGE _ MNG thread: reading the analyzed WAL from the SPSC queue of the self, and then distributing the WAL to a REDO _ PAGE _ WORKER thread;
REDO _ PAGE _ WORKER thread: responsible for actual non-transactional log playback;
REDO _ TRXN _ MNG thread: reading a WAL from the SPSC queue of the WAL each time, and then allocating the WAL to a REDO _ TRXN _ WORKER thread for processing;
REDO _ TRXN _ WORKER thread: is responsible for the actual transaction log replay.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be described in detail and completely with reference to the following embodiments and accompanying drawings. It is to be understood that the embodiments described are merely illustrative of some, but not all, of the present invention and that the invention may be embodied or carried out in various other specific forms, and that various modifications and changes in the details of the specification may be made without departing from the spirit of the invention.
Also, it should be understood that the scope of the invention is not limited to the particular embodiments described below; it is also to be understood that the terminology used in the examples is for the purpose of describing particular embodiments only, and is not intended to limit the scope of the present invention.
Example 1: a method for realizing the fast keeping of the consistency of the slave computer and the host computer data under the master-slave copy mode of an OpenGauss database (see figure 1) comprises the steps of reconstructing the existing WAL playback thread, splitting the backup playback thread into multi-thread parallel playback from the single thread playback, and dividing the WAL type into a transaction log and a non-transaction log; the transaction logs include transaction operation related logs such as CSN (COMMIT SEQUENCE NUM) logs and transaction commit logs; the non-transaction logs include data operation related logs such as insert operation logs and update operation logs; aiming at the transaction log, the method still adopts independent thread playback to avoid the occurrence of out-of-order events, and aiming at the non-transaction log, the method adopts multi-thread parallel playback to accelerate the WAL playback speed.
Specifically, the method comprises the following steps:
the method comprises the following steps: the log receiving thread writes the WAL sent by the master library into a disk;
step two: reading the WAL from the disk by the Start thread, and distinguishing the WAL into a transaction log and a non-transaction log, wherein:
(a) the transaction log is played back by the REDO TRXN work thread alone,
(b) non-transactional logs are replayed in parallel by multiple threads;
step three: for the non-transaction logs, different dispatching functions are called according to the types of the non-transaction logs, the dispatching functions dispatch the non-transaction logs and place the non-transaction logs into SPSC queues of corresponding REDO _ PAGE _ WORKER threads, and if the queues are full, the start threads wait circularly;
step four: the REDO _ PAGE _ WORKER thread reads the non-transaction log from the SPSC queue and plays back the non-transaction log, and updates the position of the non-transaction log played back by the thread;
step five: and acquiring the minimum value of the non-transaction logs played back by the REDO _ PAGE _ WORKER thread by the Start thread, judging whether the transaction logs can be played back or not, and calling a corresponding playback function if the transaction logs can be played back.
Example 2: when the method is used for processing non-transaction logs by adopting multi-thread parallel playback, if dependency exists between the non-transaction logs, the dependency is divided into intra-transaction dependency and inter-transaction dependency, and for different types of dependency, the following different processing methods are adopted:
(1) for the intra-transaction dependency relationship, when a transaction has a WAL (for example, INSERT and then DELETE) which operates the same operation object (for example, the same table) for multiple times, the number of a working thread for replaying the log is calculated according to a digital label (which can be obtained according to the WAL information) which is obtained from the WAL information and identifies the physical position of the operation object during parallel replay, and the digital label identifying the physical position of the operation object has uniqueness, so that all non-transaction logs containing the same operation object (the same table) are replayed by the same working thread.
(2) And (3) packaging WALs played back in parallel aiming at the inter-transaction dependency relationship, storing related marking information (such as whether the logs need to be synchronized to other groups) in a packaging structure, setting WAL types and dispatching function mapping relationships in a database kernel, calling dispatching functions matched with the WALs aiming at different types of WALs, and dispatching non-transaction logs by the dispatching functions.
To illustrate this technical solution more clearly, the following examples are listed for demonstration:
the database master library operates as follows:
create table test(a int)；
insert into test values(1)；
database insertion data log:
a total of 3 WALs were generated:
XLOG _ HEAP _ INSERT: performing an insert operation;
XLOG _ STANDBY _ CSN _ COMMING: the WAL is used to indicate that the transaction is committing;
XLOG _ XACT _ COMMIT _ COMPACT: a transaction commit log.
And (4) backup playback:
(a) the type of the first WAL is RM _ HEAP _ ID (which can be obtained according to desc information in the WAL, and the type of the heat mapping is RM _ HEAP _ ID), the corresponding dispatch function is dispatch _ HEAP record (which is obtained through the mapping relationship between the WAL type and the dispatch function), and the dispatch _ HEAP record determines which REDO _ BATCH thread plays back the WAL according to whether the WAL relates to a data block change:
(1) if max _ bloc _ id > is 0, indicating that the data blocks are changed, traversing each data block, selecting grouping according to a physical file OID (relfilecode) of the data block, and ensuring that the WALs identical to the physical file OID are played back by the same working thread;
(2) if max _ block _ id <0, indicating that no data block change is involved, allocating the WAL to a plurality of working threads for parallel playback;
(b) the second WAL is of the type RM _ STANDBY _ ID, when WAL _ level > is hot _ STANDBY, the master library records a WAL of this type in the transaction commit phase, and the corresponding dispatch function is dispatch STANDBY record, which dispatches the WAL to REDO _ TRXN _ MNG thread for playback;
(c) the third WAL has the type RM _ XACT _ ID, the corresponding dispatch function is dispatchxcactrecord, and the dispatchxcactrecord first determines whether the WAL will delete the file:
(1) if the file is deleted, the WAL is allocated to all REDO _ BATCH threads for playback;
(2) if the file is not deleted, the WAL is assigned to the REDO _ TRXN _ MNG thread for playback.
The above description is only an example of the present invention, and is not intended to limit the present invention. Various modifications and alterations to this invention will become apparent to those skilled in the art. Any modification, replacement, or the like that comes within the spirit and principle of the present invention should be included in the scope of the claims of the present invention.
Claims (6)
1. A method for realizing fast keeping of slave computer data consistency with host computer data under OpenGauss database master-slave copy mode is characterized in that the method comprises the steps of reforming an existing WAL playback thread, splitting independent thread playback into multi-thread parallel playback, and dividing WAL types into transaction logs and non-transaction logs, wherein the transaction logs comprise transaction operation related logs, and the non-transaction logs comprise data operation related logs; aiming at the transaction log, the method still adopts independent thread playback to avoid the occurrence of out-of-order events, and aiming at the non-transaction log, the method adopts multi-thread parallel playback to accelerate the WAL playback speed.
2. The method of claim 1, wherein the transaction log comprises a CSN log and a transaction commit log; the non-transaction logs include insert operation logs and update operation logs.
3. The method of claim 1, wherein when a multi-thread parallel playback processing is performed on non-transaction logs, if a dependency exists between the non-transaction logs, the dependency is divided into an intra-transaction dependency and an inter-transaction dependency, and for different types of dependencies, the following different processing methods are adopted:
(1) aiming at the intra-transaction dependency relationship, when a WAL which operates the same operation object for multiple times exists in one transaction, the number of a working thread for replaying the log is calculated according to a digital label which is obtained from WAL information and used for identifying the physical position of the operation object during parallel replay, and the digital label for identifying the physical position of the operation object has uniqueness, so that all non-transaction logs containing the same operation object are replayed by the same working thread;
(2) and packaging WALs played back in parallel aiming at the inter-transaction dependency relationship, storing related marking information in a packaging structure, setting WAL types and dispatching function mapping relationships in a database kernel, calling the matched dispatching functions aiming at different types of WALs, and processing the inter-transaction dependency relationship by the dispatching functions.
4. The method according to claim 3, wherein the processing method (2) calls dispatch functions matched with the WALs of different types in the inter-transaction dependency relationship for the WALs of different types, and the dispatch functions process the inter-transaction dependency relationship, and specifically includes the following processing modes:
(1) when the type of the WAL is RM _ HEAP _ ID, the corresponding dispatch function is dispatchheapcord, which decides by which REDO _ BATCH thread the WAL is played back according to whether the WAL involves a data block change:
(a) if max _ bloc _ id > is 0, indicating that the data blocks are changed, traversing each data block, selecting grouping according to the physical file OID of the data block, and ensuring that the WALs same as the physical file OID are played back by the same working thread;
(b) if max _ block _ id <0, indicating that no data block change is involved, allocating the WAL to a plurality of working threads for parallel playback;
(2) when the WAL is of the type RM _ STANDBY _ ID, the corresponding dispatch function is DispatchStandbyRecord which dispatches the WAL to a REDO _ TRXN _ MNG thread for playback;
(3) when the type of the WAL is RM _ XACT _ ID, the corresponding dispatch function is DispatcXactRecord, which first determines whether the WAL will delete the file:
(a) if the file is deleted, the WAL is allocated to all REDO _ BATCH threads for playback;
(b) if the file is not deleted, the WAL is assigned to the REDO _ TRXN _ MNG thread for playback.
5. The method of claim 1 for rapidly maintaining the slave computer data consistency with the master computer data in the OpenGauss database master-slave replication mode, wherein the method comprises the following steps:
(1) the log receiving thread writes the WAL sent by the master library into a disk;
(2) reading the WAL from the disk by the Start thread, and distinguishing the WAL into a transaction log and a non-transaction log, wherein:
(a) the transaction log is played back by the REDO TRXN work thread alone,
(b) non-transactional logs are replayed in parallel by multiple threads;
(3) for the non-transaction logs, different dispatching functions are called according to the types of the non-transaction logs, the dispatching functions dispatch the non-transaction logs and place the non-transaction logs into SPSC queues of corresponding REDO _ PAGE _ WORKER threads, and if the queues are full, the start threads wait circularly;
(4) the REDO _ PAGE _ WORKER thread reads the non-transaction log from the SPSC queue and plays back the non-transaction log, and updates the position of the non-transaction log played back by the thread;
(5) and acquiring the minimum value of the non-transaction logs played back by the REDO _ PAGE _ WORKER thread by the Start thread, judging whether the transaction logs can be played back or not, and calling a corresponding playback function if the transaction logs can be played back.
6. Use of the method of any of claims 1 to 5 for enabling a slave to quickly maintain data consistency with a master in a master-slave replication mode of an OpenGauss database in a relational database management or operating system.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111592861.XA CN114281833B (en) | 2021-12-23 | 2021-12-23 | Method for realizing quick maintenance of data consistency between slave machine and host machine under OpenGauss database master-slave copy mode |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111592861.XA CN114281833B (en) | 2021-12-23 | 2021-12-23 | Method for realizing quick maintenance of data consistency between slave machine and host machine under OpenGauss database master-slave copy mode |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114281833A true CN114281833A (en) | 2022-04-05 |
| CN114281833B CN114281833B (en) | 2024-09-06 |
Family
ID=80874967
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111592861.XA Active CN114281833B (en) | 2021-12-23 | 2021-12-23 | Method for realizing quick maintenance of data consistency between slave machine and host machine under OpenGauss database master-slave copy mode |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114281833B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116302699A (en) * | 2023-03-20 | 2023-06-23 | 北京优炫软件股份有限公司 | Control method and control system for parallel playback of databases |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030061537A1 (en) * | 2001-07-16 | 2003-03-27 | Cha Sang K. | Parallelized redo-only logging and recovery for highly available main memory database systems |
| US20130117236A1 (en) * | 2011-11-07 | 2013-05-09 | Sap Ag | Database Log Replay Parallelization |
| CN103729442A (en) * | 2013-12-30 | 2014-04-16 | 华为技术有限公司 | Method for recording event logs and database engine |
-
2021
- 2021-12-23 CN CN202111592861.XA patent/CN114281833B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030061537A1 (en) * | 2001-07-16 | 2003-03-27 | Cha Sang K. | Parallelized redo-only logging and recovery for highly available main memory database systems |
| US20130117236A1 (en) * | 2011-11-07 | 2013-05-09 | Sap Ag | Database Log Replay Parallelization |
| CN103729442A (en) * | 2013-12-30 | 2014-04-16 | 华为技术有限公司 | Method for recording event logs and database engine |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116302699A (en) * | 2023-03-20 | 2023-06-23 | 北京优炫软件股份有限公司 | Control method and control system for parallel playback of databases |
| CN116302699B (en) * | 2023-03-20 | 2024-02-06 | 北京优炫软件股份有限公司 | Control method and control system for parallel playback of databases |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114281833B (en) | 2024-09-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR100926880B1 (en) | Data replication method and system in DVMS | |
| US9723097B2 (en) | Assigning shared catalogs to cache structures in a cluster computing system | |
| US7103619B1 (en) | System and method for automatic audit data archiving within a remote database backup system | |
| US6983295B1 (en) | System and method for database recovery using a mirrored snapshot of an online database | |
| EP2474919A1 (en) | System and method for data replication between heterogeneous databases | |
| US7263537B1 (en) | System and method for creating multiple QUIESCE database copies at a single server | |
| JPH1097451A (en) | Method and device for optimizing log file of client/server computer system | |
| US10445295B1 (en) | Task-based framework for synchronization of event handling between nodes in an active/active data storage system | |
| US8015375B1 (en) | Methods, systems, and computer program products for parallel processing and saving tracking information for multiple write requests in a data replication environment including multiple storage devices | |
| US7194486B2 (en) | Method and system for data processing with data replication for the same | |
| CN114281833A (en) | Method for rapidly keeping consistency of slave computer and host computer data under master-slave copy mode of OpenGauss database | |
| CN116521083A (en) | Storage method, device and medium for distributed storage cluster data | |
| CN112417043B (en) | Data processing system and method | |
| AU2015265595B2 (en) | System and method for dynamic collection of system management data in a mainframe computing environment | |
| AU2015265599B2 (en) | System and method for the production of job level pre-processed backup of critical data and/or datasets in a mainframe computing environment | |
| CN102576294A (en) | Storage system, method, and program, comprising a plurality of storage devices | |
| EP2135166B1 (en) | Service workload identification in a data storage system | |
| CN113391933A (en) | Method for processing fund | |
| US7613888B2 (en) | Maintain owning application information of data for a data storage system | |
| CN109271109A (en) | A kind of date storage method and HBA card | |
| JP2005293323A (en) | Differential distribution record reflection system between different type rdbms | |
| CN112559243B (en) | Data snapshot method and device, electronic equipment and computer readable storage medium | |
| CN108701071B (en) | High throughput, high reliability data processing system | |
| CN111159292A (en) | Data storage method and device | |
| CN113986882A (en) | Online migration method for distributed database data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |